Julia M. Rohrer
banner
dingdingpeng.the100.ci
Julia M. Rohrer
@dingdingpeng.the100.ci
Personality psych & causal inference @UniLeipzig. I like all things science, beer, & puns. Even better when combined! Part of http://the100.ci, http://openscience-leipzig.org
about effects one would expect for multilingualism. It's just that it's something else than a randomized controlled trial explicitly manipulating (as opposed to, e.g. encouraging) multilingualism which seems...rather elusive.
November 11, 2025 at 2:06 PM
I think that's a cool study! But I think an 11-week language course is a slightly different exposure than multilingualism (regular use of more than one language). I think one could make the point that these should target similar mechanisms and so the study is informative
November 11, 2025 at 2:06 PM
Reposted by Julia M. Rohrer
I think the general lessons are:

1. Be clear about your goal (_estimand_)

2. Be clear about your analytical method (_estimator_)

3. Be clear about the assumptions that connect method to goal (_design_)
November 11, 2025 at 11:24 AM
Independent replication
November 11, 2025 at 12:32 PM
Reposted by Julia M. Rohrer
I think likening that to a superposition of quantum states doesn't work. In the quantum case, make a measurement and you will end up either with one or the other. What you are describing is worse: it starts out and will remain inconsistent. Let's call it the Whitman approach to causal inference!
November 11, 2025 at 11:52 AM
With that mindset, every cause is a manipulable cause 💪🏼
November 11, 2025 at 12:29 PM
spoken like a true internet-multilingual
November 11, 2025 at 9:59 AM
😭😭😭 you can't make a joke like that while I'm handling manuscripts that essentially implement some variation of this
November 11, 2025 at 9:58 AM
For sale: randomized controlled trial, never conducted
November 11, 2025 at 9:47 AM
we just need parents with bolder scientific mindsets
November 11, 2025 at 9:47 AM
This slide unfortunately generalizes well 🥲
November 11, 2025 at 9:25 AM
0.1% of authors get punished by actually having to conduct they studies they claim will fix the issues of their study 😭
November 11, 2025 at 9:09 AM
But they do discuss it as a limitation to *check notes* their ability to link variability in specific factors such as age of acquisition and proficiency level to aging outcomes 😭
November 11, 2025 at 9:08 AM
"I have a randomized controlled trial but he goes to a different school"
November 11, 2025 at 8:37 AM
Whenever I get to review a paper that vaguely points towards the possibility of randomized experiments, I do ask them to spell out what they actually mean. If you genuinely believe that this is a solution, surely your readers deserve more than just boilerplate.
November 11, 2025 at 8:32 AM
November 11, 2025 at 8:30 AM
Fun fact in some parts of Leipzig (such as our previous apartment) it is quite easy to get fiber optic internet access because infrastructure created in the GDR was re-animated.
November 10, 2025 at 10:02 AM
The internet thing came so quickly, we barely had enough time to prepare.
November 10, 2025 at 8:56 AM
Yeah PDF is fine. Also, somewhat delightfully, our University blocks all e-mails with rtf attachements, so we cannot even collaborate with the template provided by the DFG 👌🏼
November 10, 2025 at 8:44 AM
ngl, must have been like close to 30 years ago. But the name has been etched into my memory.
November 10, 2025 at 8:28 AM
(bread looks excellent tho 👌🏼)
November 10, 2025 at 8:06 AM
A friend just forwarded me this with the words "Andrew Heiss needs to be stopped" and I couldn't agree more.
November 10, 2025 at 8:05 AM
Wilhelma!! Now that's a name I haven't heard in a long time. Leipzig only has Guanakos :(
November 10, 2025 at 8:03 AM
Even more fun is the factor analysis version. Does the factor load onto the items or do the items load onto the factor?
November 10, 2025 at 8:02 AM